ES2868189T3 - Solid oxide fuel cell with metal support - Google Patents

Abstract

A process for forming a metal-supported solid oxide fuel cell, the process comprising the steps of: a) applying a green anodic layer including nickel oxide, copper oxide and a rare earth doped ceria to a metal substrate stainless steel, where nickel oxide, copper oxide and rare earth doped ceria are powdered, the powders having a d90 particle size distribution in the range 0.1 mm to 4 mm and where a proportion by weight of nickel oxide to copper oxide is in the range of 20: 1 to 4: 1; b) compressing the anode layer in green; c) sintering the green anodic layer to form a compound that includes oxides of nickel, copper and a rare earth doped ceria; d) providing an electrolyte; and e) providing a cathode.

Description

DESCRIPTION

Solid oxide fuel cell with metal support

Countryside

The invention relates to a metal-supported solid oxide fuel cell (SOFC), processes for forming the fuel cell, and uses thereof. The invention relates in particular to metal supported solid oxide fuel cells in which the anode contains nickel and copper.

Background

A SOFC is an electrochemical device for the generation of electrical energy through the electrochemical oxidation of a combustible gas (usually hydrogen-based). The device is generally ceramic-based, using an oxygen-ion conductive ceramic derived from metal oxide as its electrolyte. Since most oxygen ion ceramic conductors (for example, doped zirconium oxide or doped cerium oxide) only demonstrate technologically relevant ionic conductivities at temperatures above 500 ° C (for electrolytes based on oxide cerium) or at 600 ° C (for zirconium oxide-based ceramics), SOFCs operate at elevated temperatures.

Like other fuel cells, SOFCs include an anode where fuel is oxidized and a cathode where oxygen is reduced. These electrodes must be able to catalyze electrochemical reactions, be stable in their respective atmospheres at operating temperature (reducing on the anode side, oxidizing on the cathode side), and be able to conduct electrons so that the electrical current generated by electrochemical reactions it can move away from the electrode-electrolyte interface.

Finding materials with the appropriate combination of properties for the anode has proven difficult, despite numerous investigations. For many years, the state of the art of the SOFC anode has consisted of a structure composed of porous metal-ceramic (cermet), with nickel as the metallic phase and an electrolytic material (usually zirconia stabilized with yttrium or scandium oxide). as the ceramic phase, although electrolytic materials based on doped ceria, such as gadolinia or ceria doped with samaria, have also been used less frequently. In this structure, nickel plays the role of catalyst, and the nickel volume fraction is significant enough that a contiguous metallic lattice is formed, thus providing the necessary electronic conductivity. The electrolyte material forms a ceramic backbone contiguous to the anode, providing a mechanical structure, improving the bond between the anode and the electrolyte, and also expanding the anode-electrolyte interfacial region at some distance from the anode.

A well known limitation of these cermet anodes is that, at cell operating temperature, the metallic nickel in the anode is only stable in a reducing atmosphere. This atmosphere is normally provided by combustible gases, so under normal operating conditions the anode is stable. However, if the fuel gas supply is interrupted with the SOFC at operating temperature, the atmosphere within the anode will become oxidizing. Under these conditions, the metallic nickel will oxidize to nickel oxide. This oxidation is associated with an increase in volume greater than approximately 40%, since the metallic nickel that has been formed by the reduction of the sintered nickel oxide does not oxidize again to the same morphology as the original nickel oxide from which it is derived. formed. Instead, it generates mesoporosity, occupying a larger volume than the original nickel oxide. This volume change in reoxidation can generate high stresses in the anode structure, which in turn can cause the anode to crack and the possible destruction of the SOFC cell.

The inability of many SOFC cells to undergo multiple reduction-oxidation (REDOX) cycles without suffering such damage has been a major factor inhibiting the widespread commercial adoption of SOFC technology for power generation, as the systems SOFCs generally require the presence of complex and expensive purge gas systems to maintain a reducing atmosphere over the anodes in the event of an unexpected fuel outage, for example due to a failure in another part of the system requiring an emergency shutdown of the system. for security reasons.

The problem of inadequate REDOX stability is particularly acute in anode-supported fuel cells, which is currently the most common form of SOFC cells. Anodic support is beneficial as it allows a very thin layer (<20mm) of electrolyte (such as stabilized zirconia) to be used, as the electrolyte is not structural. This, in turn, allows operation at a lower temperature range than electrolyte-supported cells (650 ° C to 800 ° C instead of 850 ° C to 1,000 ° C). Since the resistance of the electrolyte to the transport of oxygen ions is inversely proportional to the thickness of the electrolyte, in electrolytically supported fuel cells, the resistance caused by the thickness of the electrolyte layer is overcome by increasing the operating temperatures, taking advantage of the exponential drop in resistance with temperature. By being able to use thinner layers in anode-supported cells, operating temperatures can be lowered, which is generally desirable as it facilitates the use of lower cost materials in the SOFC fuel cell system, and reduces the rate. of diverse mechanisms of degradation of the materials, such as the oxidation of the metallic components.

Despite these advantages, as the anode is the structural support of the SOFC cell in an anode-supported cell, the cells are very prone to catastrophic failure on repeated REDOX cycles, as stress-induced cracking can result in rupture. complete cell.

Despite considerable efforts by developers, no nickel alternative has achieved widespread adoption, as no suitable material has yet been developed that combines the relatively low cost of nickel, high catalytic activity for both the electrochemical oxidation of nickel. hydrogen as for steam reforming of hydrocarbon fuels, and high electronic conductivity.

Gorte et al. (US 2005/227133 A1, US 7,014,942 B2) have reported the use of copper in a SOFC anode partially or totally replacing nickel. Copper has advantages as an electronic conductive phase at the anode, in particular that it does not catalyze the formation of carbon from hydrocarbons. However, it is a poor catalyst for the electrochemical oxidation of hydrogen and the steam reforming of hydrocarbon fuels, which is why in the copper anodes tested by Gorte et al. An additional catalyst, such as ceria, was necessary to achieve adequate electrode performance. The other problem with the use of copper in conventional SOFC applications is that both metallic copper and copper oxide have low melting points (1,084 ° C and 1,326 ° C respectively). Cermet anodes are typically formed by sintering a mixture of metal oxide powder and electrolyte powder at 1,200 ° C - 1,500 ° C in air, followed by reduction of metal oxide to metal using hydrogen in the first SOFC run. This sintering temperature range is close to or above the melting point of copper oxide (nickel oxide, on the other hand, melts at 1,955 ° C), causing excessive sintering of the copper oxide phase . In addition, conventional operating temperatures for SOFCs are in the 700 ° C - 900 ° C range, close to the melting point of metallic copper, which tends to cause sintering of the copper phase during operation of the copper. SOFC, which can cause performance degradation. To solve this problem, Gorte et al. developed a method to add copper to the anode in a post-sintering infiltration stage, using solutions of copper salts that were dried and then calcined to break down the salt into copper oxide, thus avoiding the need to sinter the copper oxide at high temperatures. However, the infiltration stage, while allowing the use of copper cermets, can be difficult to scale to industrial production. Another problem with copper is that, although it is less reactive than nickel, it continues to oxidize if exposed to an oxidizing atmosphere at temperature, so a copper-based anode also lacks REDOX stability.

There are factors related to the design of the SOFC that can help mitigate the detrimental effects of REDOX cycles; These include:

• Do not use a cell with anodic support - the anode may therefore be thinner, reducing the total volume change through REDOX cycles and the danger of catastrophic cracking.

• Operate at a lower temperature - the oxidation rate of nickel increases exponentially with increasing temperature, starting at> 300 ° C. The lower the operating temperature, the lower the risk of nickel oxidation and volume expansion. Furthermore, nickel particles tend to oxidize through a core and shell mechanism, in which the outer surface oxidizes rapidly, but the core of the particle oxidizes more slowly as it is limited by diffusion. Therefore, at lower temperatures, it is likely that only the outer surface of the nickel particles will be reoxide at the anode, not the entire particle, and any change in volume will be small.

• Provide the anode with a contiguous ceramic "backbone" - Since the electrolyte-based ceramic phase used in SOFC cell anodes is largely unaffected by changes in oxygen partial pressure, this part The anode will not change in volume during REDOX cycles that affect the nickel phase. Therefore, the structural integrity of the anode and its bond to the electrolyte will be enhanced if there is a sintered porous ceramic network within the anode.

One SOFC cell design that has the potential to meet these criteria is the metal supported SOFC design described by the Applicant in GB 2 368 450. This SOFC cell uses a ferritic stainless steel sheet as a structural support. The foil is made porous in its central region to allow fuel access to the anode. The active layers of the cell (anode, electrolyte and cathode) are deposited on the substrate sheet as films. This means that the anode only needs to be approximately 15mm thick, as it is not the structural support of the cell. This cell is also capable of operating at temperatures in the range of 450 ° C to 650 ° C, much lower than standard operating temperatures. This is achieved through the use of ceramic materials based predominantly on cerium oxide (ceria) such as CGO10 (cerium oxide doped with gadolinium, with CGO 10 - Ce0.9Gd0.1O1.95) as an oxygen ion conducting electrolyte, which have an inherently higher oxygen ion conductivity than zirconia-based materials. To avoid internal short-circuit of the cell due to the mixed ionic-electronic conductivity of the ceria-based electrolytes, a thin film of stabilized zirconia is deposited on the electrolyte, as described in GB 2 456 445, but as the layer Zirconia is so fine, its resistance to oxygen ion transport is low enough not to impede low temperature operation. The SOFC cell of GB 2 368450 uses a cermet anode composed of porous metal and CGO10 manufactured as a thick film with a thickness of between 5 mm and 30 mm. The anode is generally deposited by screen printing an ink containing metal oxide and CGO10 powders and a porous ceramic layer is formed by a thermal process to sinter the deposited powders and form a contiguous structure bonded to the steel substrate.

A limitation imposed by the deposition of ceramic layers on a ferritic stainless steel support by conventional ceramic processing methods is the maximum temperature to which the steel can be exposed in an oxidizing atmosphere due to the formation of an oxide scale. chromium at high temperatures in an oxidizing atmosphere. This upper limit is substantially below the 1,200 ° C - 1,500 ° C typically used when sintering ceramics, so methods have been developed to sinter rare earth doped ceria electrolytes to> 96% of the theoretical density at <1,100 ° C, facilitating the formation of the desired gas tight layer (GB 2368450, GB 2386 126 and GB 2400486).

Surprisingly, sintering an anode made of nickel oxide and rare earth doped ceria at these temperatures has proven more difficult than sintering the electrolyte. This is because composites of two different oxide materials have been found to sinter worse than a single phase material. Thus, nickel oxide or ceramic alone will sinter adequately at these temperatures, but sintering of a compound in air can be poor, resulting in weak connections between the particles and a weak ceramic structure. This can lead to cell failure as a result of the REDOX cycle, as the weak bonds between the nickel particles are broken as a result of volume changes during the REDOX cycle. Ultimately, this can lead to catastrophic cell failure due to detachment of the electrolyte from the anode.

To improve the REDOX stability of the cell, it is desirable to find a means that allows sufficient sintering of the cermet structure in the temperature range in which it is possible to fire the ceramic layers on a steel substrate. Therefore, it would be advantageous to provide a method of preparing a metal-supported SOFC in which the anode is stable with redox cycles, robust to loss of reducing atmosphere at operating temperature, and yet capable of manufactured using commercially viable production methods. The invention aims to overcome or improve at least some aspects of this problem and those described above.

Summary

Consequently, in a first aspect of the invention a process for forming a metal-supported SOFC is provided, the process comprising the steps of:

a) apply a green anodic layer that includes nickel oxide, copper oxide and a rare earth doped ceria to a stainless steel metal substrate where nickel oxide, copper oxide and rare earth doped ceria are pulverized, the powders having a d90 particle size distribution in the range of 0.1 mm to 4 mm and wherein a weight ratio of nickel oxide to copper oxide is in the range of 20: 1 to 4: 1;

b) compressing the anode layer in green;

c) sintering of the green anodic layer to form a compound that includes nickel oxides, copper oxides and a rare earth doped ceria;

d) providing an electrolyte; and

e) providing a cathode.

The presence of copper in the anode layer, usually as copper oxide, provides an anode with better sintering between nickel oxide and rare earth doped ceria. This, in turn, improves the formation of the ceramic backbone at the anode and increases the stability of the anode (and the fuel cell as a whole) against REDOX cycles, since the microstructure of the anode is more robust than if copper were absent and less likely to change in volume during the reduction of nickel and copper oxide to nickel and copper on the first use of the fuel cell, or during any change in reoxidation if the reducing atmosphere is lost to operating temperatures, for example in the event of unforeseen system failure and loss of fuel supply.

The process of the invention will further comprise the step of compressing the anode layer in green, often at pressures in the range of 100 to 300 MPa. This compression step increases the density of the unsintered green anode layer, ensuring that the rare earth doped nickel oxide, copper oxide and ceria particles are in close enough contact to effectively sinter at the temperatures employed in the process of the invention. Often the compression stage is used in combination with a printed layer heating stage to remove residual organic materials from the ink base prior to compression, to leave a green anodic layer containing nickel oxide, oxide copper and a rare earth doped ceria that can be compressed.

The first step of the described process is the application of a layer of anode in green to the metallic substrate; the metal substrate will be a stainless steel substrate, in particular a ferritic stainless steel substrate, since the steel Ferritic stainless forms a surface passivation layer of chromium oxide when heated. This passivation layer protects the thick metal of the support and provides a diffusion barrier between the anode and the thick metal of the support. As used herein, the terms "support" and "substrate" to refer to the metallic support / substrate are used interchangeably. The formation of a passivation layer of chromium oxide, unlike aluminum oxide or silicon oxides that are usually formed with other heat-resistant steels, has the advantage that chromium oxide is an electronic semiconductor at high temperatures, rather than being insulating, making ferritic stainless steel suitable for use in fuel cell applications. The ferritic stainless steel can be a ferritic stainless steel without aluminum, such as a ferritic stainless steel containing titanium and / or niobium as stabilizers. Often times, ferritic stainless steel will comprise between about 17.5 and 23% by weight of Cr. In particular, ferritic stainless steel can be selected from the European designation 1.4509 (17.5 to 18.5% by weight Cr) and / or the European designation 1.4760 (22 to 23% by weight Cr), although similar ferritic stainless steel designations may also be used, as will be understood by one of ordinary skill in the art.

The substrate may have a thickness in the range of about 50mm to 500mm, often from about 100mm to 400mm, in some cases from about 200mm to 350mm. The thickness of the substrate is determined by the need to provide a stable substrate, which does not warp during cell formation or in use, but which is as thin as possible to allow efficient contact between the fuel and the anode. As described in GB 2368450, this contact can be achieved with excellent results by providing a porous region delimited by a non-porous region of the substrate, on which the anode is formed. It will often be the case that the porous region of the substrate includes a plurality of passage openings that fluidly interconnect one and the other surfaces of the substrate, and these will often be evenly spaced, additionally or alternatively having a lateral dimension of approximately 5 mm. to 500mm, or approximately 100mm to 300mm. Furthermore, the apertures may comprise from about 0.1 to 5% of the area of the porous region of the substrate or from about 0.2 to 2% of the area of the porous region of the substrate. Each of these features contributes to efficient fuel transfer through the substrate to the anode, while allowing the metal substrate to support the fuel cell, facilitating the use of dramatically reduced thicknesses of the electrochemically active layers within the cell. .

Typically the substrate will be a sheet, although a sintered substrate could also be used. The advantage of the sheets is the ease of control of the structure of the porous region.

The green anode layer is generally formed by applying an ink comprising nickel oxide, copper oxide, and rare earth doped ceria, although other methods can be used. These three components will generally be suspended as powders within an ink base, the ink base generally containing one or more volatile solvents, one or more dissolved non-volatile polymeric binders, dispersants, wetting agents, and other common ink components. Nickel oxide, copper oxide and rare earth doped ceria will have a d90 particle size distribution in the range of 0.1 mm to 4 mm, often 0.2 mm to 2 mm or 0, 7mm to 1.2mm. Although the particle size distributions, and the sizes themselves, of each of the copper oxides, nickel oxides, and rare earth doped ceria may be different, it may be beneficial if they are the same, or similar, as this helps. to facilitate a good mixing of the powders and, therefore, a robust sintering of the anode. In general, small size particles are selected, since these are more easily suspended in the ink and offer a greater homogeneity of the components within the anode layer, in addition to having a higher surface / volume ratio, which increases the particle reactivity and ease of sintering.

Typically, the ink will contain in the range of 30 to 70% by weight of the solids content in the ink of mixed metal oxides (namely, the combination of copper oxide and nickel oxide). Often this will be 35 to 45% by weight with the remainder of the solids being the rare earth doped ceria. It must be said that it will often be the case that the only solids in the ink will be the metallic oxides and the rare earth-doped ceria, and therefore the anode will often be composed, or essentially composed, of oxide nickel, copper oxide and ceria doped with rare earths. Often times, the metal oxide component of the ink will contain a range of 5 to 50% by weight of the total metal oxide of copper oxide, often 8 to 25% by weight. In many cases, the copper oxide will be around 10% by weight, perhaps between 8 or 9 and 11 or 12% by weight of the total metal oxide, so the ratio of nickel oxide to oxide copper is approximately 9: 1. The ratio is in the range of 20: 1 to 4: 1, often in the range of 15: 1 to 6: 1. By selecting these levels of copper oxide doping, it has been found that the relatively low melting point of copper oxide offers improved sintering within the composite anode material, without lowering the melting point of the metal oxide below that necessary to that the sintering of the rare earth doped ceria occurs, and without impairing the performance of the anode, particularly when the fuel is hydrogen, or when the fuel cell vaporizes the hydrocarbons.

It will often be the case that copper oxide is copper (II) oxide, as it has a higher melting point than copper (I) oxide and has semiconductor properties. However, copper (I) oxide can also be used, as this can form copper (II) oxide at high temperatures in air.

In many examples, the rare earth doped ceria will have the formula Ce1-xRExO2-x / 2, where RE is a rare earth and 0.3>x> 0.05. Often the rare earth doped ceria will be gadolinium doped cerium oxide, often of formula Ce0.9Gd0, iOi, 95 (CGO10). These compounds are generally used because they have higher oxygen ion conductivity than many electrolyte materials, including zirconia-based materials, allowing fuel cell operation at lower temperatures than conventional SOFC fuel cells. , the operating temperature of the fuel cell of the invention being normally in the range of 450 ° C to 650 ° C, often 500 ° C to 620 ° C. Operating the fuel cell at lower temperatures has a number of advantages, including reducing the rate of nickel oxidation in non-reducing atmospheres, which in turn often results in only the outer shell of the fuel cell being oxidized. the particle, reducing the volume change within the anode and therefore the risk of cracking in the event that the reducing atmosphere of the fuel supply is interrupted. In addition, it makes it possible to use metal supports, allowing thinner layers of electrode material and electrolyte to be used, as these play a minor, if any, structural role. Furthermore, these temperatures are well below the melting point of copper, which offers the possibility of using copper as a component of the cell.

The application of the green anodic layer generally includes an initial application of the ink to the metallic substrate, which will normally be by printing, for example by screen printing, although other methods can be used, such as tape casting, vacuum casting. , electrophoretic deposition and calendering, as will be known to the person skilled in the art. When there is a porous region, the application of the ink to the substrate is usually such that a layer is formed over the porous region, but the non-porous region is left substantially uncovered. This ensures that the fuel cannot bypass the anode, but minimizes material costs and weight by not covering more of the substrate than necessary.

This initial application will optionally be followed by a step of drying the ink to provide a printed layer. Drying can be in the air or with gentle heat. Gentle heat is often used to speed up the formation of the printed layer. Temperatures in the range of 50 ° C to 150 ° C would be typical. The drying step evaporates the solvents and fixes the binders in any ink formulation used, solidifying the ink and forming an initial, albeit brittle, anode layer, here called the printed layer. This layer will generally have a thickness in the range of 5mm to 40mm, often 7mm to 20mm, and often 9mm to 15mm. As the fuel cells of the invention are not anode-supported cells, the anode layer can be much thinner than in many conventional fuel cells, which has the advantage that the overall volume change during the REDOX cycle is less, and thus the cracking of the anode over time is significantly reduced.

Nickel oxide, copper oxide and applied rare earth doped ceria, or the printed layer if there is a drying step, can then be heated to remove any organic components in an ink mixture, for example polymeric binders normally present in inks. The temperature of this step will depend on the binders present, but will often be in the range of 300 ° C to 500 ° C. This heating step can be combined with the drying step, although to provide a well-formed, uniform, green anodic layer, the solvents are generally removed first, and then the organic components of the mixture in a separate step.

Often times, the compression step described above will be applied after the ink has dried and the organic components have been removed, since at this stage the anode layer in green contains only the active components (namely the oxide of nickel, copper oxide and rare earth doped ceria). This allows the compression stage to compact the anode in the most efficient way and increase the density of the oxides and ceria to improve sintering. Various compression methods can be used, as will be known to one of ordinary skill in the art, although uniaxial or cold isostatic pressing is often used.

The step of firing (sintering) the anode layer in green to form a compound that includes oxides of nickel, copper and a rare earth doped ceria allows sintering of the rare earth doped ceria and metal oxides to form the ceramic structure anode. Therefore, the baking of the anode layer in green is generally carried out in an oven at a temperature in the range of 950 ° C to 1,100 ° C, often from 980 ° C to 1,050 ° C or 1,000 ° C. and 1,030 ° C. The upper limit of these ranges is selected based on the stability of the substrate. Above about 1,100 ° C, even high-chromium steels, known for their high resistance to oxidation, rust in air too quickly for the substrate to survive the firing process. In particular, the chromium oxide passivation layer repeatedly grows and flakes during the formation of the anode cermet, weakening the metal substrate to an unacceptable point. The use of rare earth doped ceria facilitates the use of a metallic substrate, along with the formation of a robust cermet, as ceria compounds can sinter at temperatures below 1,100 ° C. The lower limit is driven by the need for successful sintering of the materials.

The cooking stage will normally be carried out in air, although other non-reducing atmospheres can be used. Typically the cooking stage will last between 15 and 60 minutes. While the firing period should be sufficient to allow sintering of the metal oxides and rare earth doped ceria, and to allow the furnace to reach thermal equilibrium, too long a firing period can increase oxidation of the metal support and cause contamination of the anode, when ferritic stainless steel is used, with chromium evaporating from the support. Therefore, the optimal cooking period is in the range of 15 to 60 minutes. After sintering, the anode is allowed to cool, providing a robust and porous anode structure containing rare earth doped nickel, copper and ceria as a ceramic conductor of oxygen ions.

Although, as described above, the anode firing occurs prior to supplying the electrolyte, it may be that the electrolyte is applied on the anode layer in green before firing occurs. Thus, the process may comprise the step of providing an electrolyte prior to the step of firing the green anode layer, so that the electrolyte and the green anode layer are fired simultaneously.

Typically for use with the fuel cells of the invention the electrolyte will have a thickness in the range of 5mm to 30mm, often in the range of 10mm to 20mm. The provision of such a thin electrolyte layer allows rapid transfer of oxygen ions from the cathode to the anode. Often times, the electrolyte will consist of a rare earth doped ceria, with appropriate rare earth ceria being those defined above for the anode. In some examples, the electrolyte may consist of a rare earth doped ceria combined with a low content of cobalt oxide and / or copper oxide, as sintering aids; for example, there may be 0.5 to 5% by weight of cobalt oxide and / or copper oxide, with the remaining electrolyte being rare earth doped ceria. The use of rare earth doped ceria for both the anode and the electrolyte helps to improve the compatibility between these fuel cell components both chemically and in terms of thermal expansion, which is closely related to the reduction of mechanical stress. between layers during the REDOX cycle, and therefore also reducing the probability of cracking and failure of the fuel cell in use. Furthermore, as these ceria have high charge transfer rates, their inclusion ensures a good charge transfer rate between the electrolyte and the anode.

The electrolyte will generally be sintered, either simultaneously with the anode as described above, or in a separate firing step after the anode is fully formed.

Typically the cathode will have a thickness in the range of 30mm to 60mm, often 40mm to 50mm. The cathode will generally comprise two layers, a thin active layer where oxygen reduction takes place, and a thicker current collecting layer, to allow current from one cell of the stack to be collected. The current collecting layer will generally be a perovskite such as strontium lanthanum cobaltite, although any electrically conductive ceramic material can be used.

The cathode of the active layer may comprise a mixed conductor sintered powder mixture of perovskite oxide and rare earth doped ceria, the rare earth doped ceria being defined above. Perovskite can consist of La1-xSrxCoyFe1-yO3-5, where 0.5> x> 0.2 and 1> y> 0.2. In particular, the perovskite oxide mixed conductor can comprise one or more of La0.6Sr0.4CO0.2Fe0.8O3-5, Gd0.5CoO3-5, and RExSnxCoO3-d, (where RE = La, Sm , Pr y 0.5 <x <0.8). It may be helpful to use these compounds as they have higher ionic conductivity than most perovskites. In some cases, the mixture contains between 20 and 50% by weight of rare earth doped ceria, in some cases between 30 and 45% by weight, in some cases between 35 and 45% by weight, or about 40% by weight of rare earth doped ceria, as defined above. This helps to improve the compatibility between the cathode and the electrolyte both chemically and in terms of the thermal expansion described above, and since these ceria have high charge transfer rates, their inclusion ensures a good rate of charge transfer between the electrolyte and the cathode.

The cathode will generally be sintered before use. The cathode will normally be applied as one or more layers (eg active and current collection) directly or indirectly on the sintered electrolyte and sintered under conditions similar to those described above for the anode. In this way, an intermediate temperature metallic type fuel cell is obtained, which is resistant to repeated REDOX cycles and, as a result of the anode structure formed, to high temperature fuel depravity.

In a second aspect of the invention there is provided a SOFC metal supported fuel cell comprising an anode, a cathode and an electrolyte, wherein the anode includes nickel, copper and rare earth doped ceria. As described above, to provide a robust fuel cell structure, nickel, copper, and rare earth doped ceria are generally sintered. In use, the nickel may be in a form comprising metallic nickel, nickel oxide, and combinations thereof, depending on the REDOX state of the nickel. For example, nickel will be in the nickel oxide form at the time of cell formation, but will be reduced to metallic nickel at the time of the first use of the cell. Similarly, copper can be in a form comprising metallic copper, copper (I) oxide, copper (II) oxide, and combinations thereof. Furthermore, mixed phases of metal and metal oxide can form, due to the mutual solubility of nickel and copper at high temperatures. Thus, it may be the case that an alloy of nickel and copper is formed that, when oxidized, forms a mixed metal oxide of nickel and copper, which could generally be described as having the formula NixCu1-xO, where x is variable between 0 and 1, as the expert reader will understand. Mixed oxides containing cerium could also form, due to the solubility of copper oxide in doped ceria. The anode, cathode and electrolyte will, in other respects, generally be as described above.

In some cases, the fuel cell will be a fuel cell of the type described in GB 2368450 issued by Applicants, which is incorporated herein by reference. In such cases, the fuel cell may comprise:

(i) a ferritic stainless steel support that includes a porous region and a non-porous region that limits the porous region;

(ii) a ferritic stainless steel bipolar plate located under one of the surfaces of the porous region of the support and tightly attached to the non-porous region of the support around the porous region of the support;

(iii) an anode comprising an anode layer located on the other surface of the porous region of the support;

(iv) an electrolyte comprising an electrolytic layer located on top of the anode layer; and

(v) a cathode comprising a cathode layer positioned on top of the electrolyte layer; wherein the anode includes nickel, copper and a rare earth doped ceria.

The fuel cell may be present in a fuel cell stack, comprising two or more fuel cells, and therefore provided in a third aspect of the invention, a fuel cell stack comprising fuel cells according to the second aspect of the invention. Each fuel cell can comprise a bipolar plate, as described above, to which the bracket can be welded, or otherwise sealed.

In a fourth aspect of the invention, the use of a fuel cell according to the second aspect of the invention in the generation of electrical power is also provided.

The process of the invention aims to provide a method for the manufacture of an intensely sintered rare earth doped nickel-copper-ceria thick film anode suitable for use in a SOFC metal-supported solid oxide fuel cell, avoiding at the same time the problems of a bad anodic sintering, and the delamination of the electrolyte in use. It may be the case that the process is a process to form a solid oxide fuel cell with metal support, the process comprising the steps of:

a) applying a green anodic layer including nickel oxide, copper oxide and rare earth doped ceria to a metal substrate, where nickel oxide, copper oxide and rare earth doped ceria are powdered, and the powders have a d90 particle size distribution in the range of 0.1 mm to 4 mm; optionally 0.2mm to 3mm and contain a weight ratio of nickel oxide to copper oxide in the range of 20: 1 to 4: 1, and wherein nickel oxide, copper oxide and ceria doped with rare earths are optionally applied as an ink, the ink optionally comprising a total solids content in the range of 30 to 70% of mixed metal oxides, with optionally in the range of 5 to 50% by weight of the total metal oxide of copper oxide;

b) optionally, drying the ink to obtain a printed layer having a thickness in the range of 5mm to 40mm;

c) compressing the green anode layer at pressures optionally in the range of 100 MPa to 300 MPa;

d) optionally heating the printed layer to remove the ink base leaving a green anodic layer comprising nickel oxide, copper oxide and a rare earth doped ceria;

e) sintering the green anode layer at a temperature optionally in the range of 950 ° C to 1,100 ° C to form a composite;

f) providing an electrolyte; and

g) providing a cathode.

Unless otherwise indicated, each of the integers described in the invention may be used in combination with any other integer, as would be understood by one of ordinary skill in the art. Furthermore, while all aspects of the invention preferably "comprise" the features described in connection with that aspect, it is specifically envisaged that they may "consist" or "consist essentially" of the features described in the claims. Furthermore, it is understood that all terms, unless specifically defined herein, are given their commonly understood meanings in the art.

Furthermore, in examining the invention, unless otherwise indicated, the description of alternative values for the upper or lower limit of the allowed range of a parameter should be construed as an implicit statement that each intermediate value of said parameter, which is between the smallest and largest of the alternatives, is also itself described as a possible value for the parameter.

Furthermore, unless otherwise indicated, all numerical values appearing in this application are to be understood as modified by the term "about".

Brief description of the drawings

In order that the present invention may be more easily understood, it will be described below with reference to the figures and specific examples.

Figure 1 is a schematic representation of a SOFC as described in GB 2368450;

Figure 2 is a scanning electron micrograph (SEM) showing a cross section through a SOFC of Figure 1 (15.0 kV, 7.9 mm x 1.50k);

Figure 3 is an SEM showing a cross section through a SOFC that includes an anode formed from a nickel oxide-copper oxide-GCO compound (15.0 kV, 8.5 mm x 1.0k ,);

Figure 4 is a SEM showing a cross section of a sintered anode formed from a nickel oxide-GCO compound (15.0 kV, 7.1 mm x 13k);

Figure 5 is an SEM showing a cross section of a sintered anode formed from a nickel oxide-copper oxide-GCO compound (15.0 kV, 8.9 mm x 13k);

Figure 6 is a current-voltage curve of the SOFC of Figure 3 as a function of cell operating temperature (56% hydrogen fuel - 44% nitrogen, excess air fed to the cathode);

Figure 7 is a power cycle graph of the SOFC of Figure 3; and

Figure 8 is a table showing the improved mechanical strength of the nickel oxide-copper oxide-GCO compound illustrated in Figure 3 in relation to the nickel-CGO anode illustrated in Figure 2.

Detailed description

In figure 1 a SOFC 10 is schematically shown, as described in document GB 2368450, and in cross section of SEM in figure 2. Both figures show a ferritic stainless steel substrate 1, made partially porous by the laser drilling of thousands of holes through the central region of the substrate 2. The porous substrate is covered by an anodic layer 3 of nickel oxide and CGO covering the porous region 2 of the substrate 1. On the anodic layer 3 is deposits a layer 4 of CGO electrolyte (10 mm - 20 mm, CGO), which is superimposed on anode 3 in the unperforated zone 9 of substrate 1, thus forming a seal around the edge of anode 3. The cathode 5,6 It has a thin active layer 5 (CGO compound) where oxygen reduction takes place, and a thicker current collecting layer 6 (lanthanum strontium cobaltite) to allow current collection from cell 10 of a battery. Figure 2 also shows a very thin stabilized zirconia layer 7 and an even finer doped ceria layer 8, which block the electronic conductivity (preventing short circuits due to unwanted chemical reactions between the cathode 5,6 and the zirconia layer 7 ) and form the interface between anode 3 and electrolyte 5,6 respectively.

The SOFC 10 of Figures 1 and 2 was prepared by applying a screen printing ink containing suspended particles of nickel oxide powder and CGO powder (d90 = 0.7 mm to 1.2 mm, the ratio of oxide to nickel and the CGO in the ink of 1.8: 1 by weight). The ink was screen printed onto the ferritic stainless steel substrate 1 using conventional methods, and dried in an oven to evaporate the solvents and fix the binders, thus forming a dry printed layer 9mm to 15mm thick. The dried and printed layer was compressed by cold isostatic pressing at a pressure of 300 MPa. The anode layer in green was placed in an oven and heated at a temperature of 960 ° C in an air atmosphere for 40 minutes, to produce a sintered anode layer 3. A CGO electrolyte layer 4 was sprayed onto the anode layer 3 and baked in an oven at 1020 ° C for 40 minutes. Finally, the zirconia layer 7 was applied to the fired electrolyte layer by the method described in GB 2 456 445, followed by the application of the doped ceria layer 8 and the two cathodic 5,6 layers, also using the methods of GB 2456445, prior to firing at a temperature of 825 ° C to produce structure 1 of the SOFC fuel cell.

Figure 3 shows a cross section through a SOFC including a nickel oxide-copper oxide-CGO compound according to the claim. Nickel oxide and copper oxide are present in a 9: 1 ratio by weight, resulting in a 9: 1 ratio of nickel to copper in use. Subject to the introduction of copper into the anodic structure, such that the 1: 1.3 ratio of nickel oxide: CGO described above becomes a 1: 1.3 ratio of the mixed metal oxide (namely, nickel oxide and copper oxide) relative to CGO, the fuel cell structure was in line with the prior art cell of Figures 1 and 2. The fabrication closely followed the prior art cell preparation method , with the exception that the dried printed layer was heated in an oven at a temperature of 350 ° C before compression to remove organic binders from the ink and provide a green anodic layer, and that the anode firing was performed at 1,020 ° C for 45 minutes.

Examples

Anode structure

Figures 4 and 5 show the difference in the anode structure obtained by adding copper oxide to the compound structure. The compound of Figure 4 has a composition of 64% by weight of nickel oxide and 36% by weight of CGO and the compound of Figure 5, 51% by weight of nickel oxide, 5.7% by weight of oxide from copper (II) and 43.3% by weight of CGO. To improve the REDOX stability of the nickel-copper anode in Figure 5, the metal oxide level was reduced somewhat relative to the original anode shown in Figure 4. After reduction during fuel cell operation, the The cermet of the anode in Figure 4 has 53% by volume of metal, compared to 45% by volume of metal in Figure 5. It has been shown that reducing the metal content alone does not confer an adequate level of REDOX stability. ; the addition of copper is also required. Both compounds were prepared as above, and fired in air at 1,020 ° C for 60 minutes prior to cell fabrication and metal reduction in situ to form the cermets shown.

Good sintering is evidenced by a clear distinction between the ceramic and metallic regions, and because the particles of both the ceramic and metallic phases have fused together. Ceramic regions appear as light regions and metallic regions as dark spots. As can be seen, the compound of Figure 5, which contains copper, includes larger and darker metallic particles, indicating good sintering; the well sintered structure of the CGO is also easily observable. This well sintered structure can also be seen in figure 3 (anode 3).

The resulting anode structure has been shown to be highly stable to REDOX at operating temperatures of <650 ° C, being capable of withstanding hundreds of high temperature fuel outages without significant degradation of cell performance.

Copper selection

A number of cations are known to enhance sintering of doped ceria, including copper, cobalt, iron, manganese, and lithium (US 6709628, JD Nicholas and LC De Jonghe, Solid State Ionics, 178 (2007) , 1187-1194). Therefore, the possibility of doping rare earth-doped ceria with one of these cations was considered. Of the cations mentioned, copper, cobalt and lithium are the most effective in enhancing sintering of rare earth doped ceria. Copper and cobalt are the only cations that the applicant considers suitable for use in an anode of a solid fuel cell (SOFC), as lithium oxide is highly reactive and is also known to be very harmful to the ionic conductivity of the rare earth doped ceria by forming an insulating phase at the grain boundaries. Cobalt is known to enhance the sintering of rare earth-doped ceria and is also effective as an anode catalyst (CM Grgicak, RC Green and JB Giorgi, J. Power Sources, 179 (1), 2008, 317- 328), although normally less than nickel. However, the initial evaluation of the sintering behavior of composites using a pushrod dilatometer surprisingly showed that cobalt oxide is not effective in improving the sintering of nickel oxide, and therefore the sinterability of the composite material of the anode was not significantly improved by the partial or even complete replacement of nickel oxide by cobalt oxide. Copper oxide, on the other hand, showed a large increase in the sinterability of the compound, in part it is suspected that because it can form a low-melting eutectic with nickel oxide, thus introducing some sintering in the liquid phase.

Fuel cell performance

Figure 6 is a series of current-voltage bias curves for the fuel cell of Figure 3, at different operating temperatures. The feed rate was calculated to give approximately 60% fuel utilization at 0.75V / cell at each of the measured temperatures, demonstrating that the system can operate over a temperature range as wide as at least , from 495 ° C to 616 ° C, which allows optimizing the operating temperature depending on the application, the number of cells in the stack, the required production, etc.

Figure 7 shows the very good REDOX stability possible with this anode structure. A series of cycles is run at 600 ° C in a short seven-layer cell, where a current-voltage curve is run to establish the performance of the cell. The cell is then returned to an open circuit and the hydrogen supply to the cell is cut off while keeping the cell at 580 ° C - 600 ° C. During this period, air and nitrogen are kept in the pile. The fuel shutdown is maintained for 20 minutes, allowing time for the anode to partially reoxide. The hydrogen feed is then restored, and after allowing the cell to recover for a few minutes, another current-voltage curve is run to determine if cell performance has been lost as a result of the anode REDOX cycle. This sequence continues until the performance of the stack begins to decline, indicating that one or more cells have been corrupted as a result of the REDOX cycle.

It can be seen from Figure 7 that with the SOFC cell of Figure 3, the seven cells within the stack will tolerate more than 500 REDOX cycles without any significant loss of performance after a small initial break-in, with 544 cycles executed in total.

Increased mechanical resistance of the anode thanks to the addition of copper

Figure 8 is a table of the results of the mechanical strength tests performed on the SOFC cells both after the initial manufacture and after the cells have operated in an initial performance characterization test, both for the standard nickel anodes. -CGO as illustrated in figure 2, as for CGO nickel-copper anodes as illustrated in figure 3.

In cells as they come out of manufacture, the anodes are in the oxidized state and prior to mechanical testing they are reduced to mimic the structure of the anode in the cell at the beginning of operation, while the anodes in the cells "after operation" are in the final cermet state of the working anodes.

To measure the mechanical strength of the cells, the metal substrates of the cells are first glued to a flat steel plate to prevent the cells from flexing when a tensile force is applied. The cell cathodes are mechanically removed, exposing the electrolyte.

To evaluate the mechanical strength of the anode and / or the anode-electrolyte junction, circular metallic test pieces are glued to the surface of the electrolyte at the four corners of the electrolyte and in the center of the cell. A diamond scorer is used to cut the ceramic layers of the cell around the test piece of metal. A calibrated hydraulic puller is then placed on the test piece and used to measure the tension required to extract the test piece from the cell substrate. With this technique a maximum tensile stress of 17 MPa can be applied, after which the glue holding the test piece to the electrolyte tends to fail and not the layers of the fuel cell being tested. If the test piece is shed at less than 17 MPa, this indicates the failure stress of the weakest layer of the cell (usually the internal structure of the anode).

It can be seen that while standard CGO nickel anodes are strong in their out-of-the-box state, they fail at much lower stresses after reduction of nickel oxide to nickel metal in the cell "after run." Without being bound by theory, it is believed that this is largely due to the lack of a contiguous ceramic structure within the anode, meaning that the mechanical strength of the anode is provided entirely by relatively weak connections between the nickel particles. In contrast, CGO nickel-copper anodes are observed to retain their strength upon reduction to the cermet structure, indicating much greater sintering of the metallic and ceramic phases.

It should be appreciated that the processes and the fuel cell of the invention are capable of being incorporated in the form of a variety of embodiments, only a few of which have been illustrated and described above.

Claims (14)

1. A process to form a solid oxide fuel cell with metal support, the process comprising the steps of: a) apply a green anodic layer that includes nickel oxide, copper oxide and a rare earth doped ceria to a stainless steel metal substrate, where nickel oxide, copper oxide and rare earth doped ceria are pulverized, the powders having a d90 particle size distribution in the range of 0.1mm to 4mm and wherein a weight ratio of nickel oxide to copper oxide is in the range of 20: 1 to 4: 1 ; b) compressing the anode layer in green; c) sintering the green anode layer to form a compound that includes nickel oxides, copper and a rare earth doped ceria; d) providing an electrolyte; and e) providing a cathode. 2. A process according to claim 1, wherein the compression of the green anode layer occurs at pressures in the range of 100 MPa to 300 MPa. 3. A process according to claim 1 or claim 2, wherein the sintering of the green anode layer occurs at a temperature in the range of 950 ° C to 1,100 ° C. 4. A process according to any preceding claim, wherein nickel oxide, copper oxide and rare earth doped ceria are applied as an ink. 5. A process according to claim 4, wherein the ink comprises between 5 and 50% by weight of the total metal oxide of copper oxide. A process according to claim 4 or 5, wherein the application of the green anodic layer includes an initial application of the ink to the metallic substrate, and drying of the ink to provide a printed layer of a thickness in the range of 5mm to 40mm. A method according to any preceding claim, further comprising heating the printed layer to remove the base ink leaving a green anodic layer comprising nickel oxide, copper oxide and a rare earth doped ceria. A process according to any preceding claim, wherein the step of providing an electrolyte occurs prior to the step of sintering the green anode layer, so that the electrolyte and the green anode layer are sintered simultaneously. 9. A metal supported solid oxide fuel cell obtained by the process according to any preceding claim. 10. A fuel cell according to claim 9, wherein the nickel is in a form selected from metallic nickel, nickel oxide, a nickel-copper alloy, a nickel-copper oxide, and combinations thereof. A fuel cell according to claim 9 or claim 10, wherein the copper is in a form selected from metallic copper, copper (II) oxide, copper (I) oxide, a nickel-copper alloy , a nickel-copper oxide, and combinations thereof. 12. A fuel cell according to any of claims 9 to 11, comprising (i) a ferritic stainless steel support that includes a porous region and a non-porous region that limits the porous region; (ii) a ferritic stainless steel bipolar plate positioned under a surface of the porous region of the support and tightly bonded to the non-porous region of the support around the porous region of the support; (iii) an anode comprising an anode layer located on the other surface of the porous region of the support; (iv) an electrolyte comprising an electrolytic layer located on top of the anode layer; and (v) a cathode comprising a cathode layer positioned on top of the electrolyte layer; wherein the anode includes nickel, copper and a rare earth doped ceria. A fuel cell according to any one of claims 9 to 12, wherein the rare earth doped ceria comprises gadolinium doped cerium oxide. 14. A fuel cell stack comprising two or more fuel cells according to any one of claims 9 to 13; or the use of a fuel cell according to any one of claims 9 to 13, in the generation of electrical energy.